1. Field of the Invention
The present invention relates to respiratory flow measurement, and specifically to improving the performance of differential pressure flowmeters under diverse inlet conditions through an improved sensor configuration.
2. State of the Art
Respiratory flow measurement during the administration of anesthesia, in intensive care environments and in monitoring the physical condition of athletes and other individuals prior to and during the course of training programs provides valuable information for assessment of pulmonary function and breathing circuit integrity. Many different technologies have been applied to create a flowmeter that meets the requirements of the critical care environment. Among the flow measurement approaches which have been employed are:
1) Differential Pressure--measuring the pressure drop or differential across a resistance to flow.
2) Spinning Vane--counting the revolutions of a vane placed in the flow path. PA1 3) Hot Wire Anemometer--measuring the cooling of a heated wire due to airflow passing around the wire. PA1 4) Ultrasonic Doppler--measuring the frequency shift of an ultrasonic beam as it passes through the flowing gas. PA1 5) Vortex Shedding--counting the number of vortices that are shed as the gas flows past a strut placed in the flow stream. PA1 6) Time of Flight--measuring the arrival time of an impulse of sound or heat created upstream to a sensor placed downstream.
Each of the foregoing approaches has various advantages and disadvantages, and an excellent discussion of most of these aforementioned devices may be found in W.J. Sullivan; G.M. Peters; P.L. Enright, M.D.; "Pneumotachographs: Theory and Clinical Application,"Respiratory Care, July 1984, Vol. 29-7, pp. 736-49, and in C. Rader, Pneumotachography, a report for the Perkin-Elmer Corporation presented at the California Society of Cardiopulmonary Technologists Conference, Oct. 1982.
At the present time, the most commonly employed device for respiratory flow measurement is the differential pressure flowmeter. Because the relationship between flow and the pressure drop across a restriction or other resistance to flow is dependent upon the design of the resistance, many different resistance configurations have been proposed. The goal of all of these configurations is to achieve a linear relationship between flow and pressure differential. It should be noted at this point that the terms "resistance" and "restriction" as applied herein to the physical configuration which produces a pressure drop or differential for use as a flowmeter input signal may be used interchangeably.
In some prior art differential pressure flowmeters (commonly termed pneumotachs) the flow restriction has been designed to create a linear relationship between flow and differential pressure. Such designs include the Fleisch pneumotach in which the restriction is comprised of many small tubes or a fine screen, ensuring laminar flow and a linear response to flow. Another physical configuration is a flow restriction having an orifice variable in relation to the flow. This arrangement has the effect of creating a high resistance at low flows and a low resistance at high flows. Among other disadvantages, the Fleisch pneumotach is susceptible to performance impairment from moisture and mucous, and the variable orifice flowmeter is subject to material fatigue and manufacturing variabilities.
U.S. Pat. No. 5,038,773 discloses a differential pressure flowmeter sensor which employs a plurality of pressure ports or apertures symmetrically disposed on the leading and trailing edges of hollow cruciform ribs divided to define two sets of lumens and extending across the cross-section of a tubular housing. U.S. Pat. No. 5,088,332 discloses a differential pressure flowmeter sensor having first and second pressure ports or apertures axially disposed within a tubular housing and supported therein by longitudinally-extending vanes or baffles including surfaces thereon for collecting and guiding pressure generated by gas flowing in the housing to the pressure ports. The flowmeter designs of the foregoing patents are intended to address deficiencies in other prior art flowmeter sensors with regard to performance impairment due to moisture and mucous, and to provide a simple design permitting economical manufacture and, if desired, disposability.
All of the prior art flowmeter sensors referenced above, however, are susceptible to performance impairment and inaccuracies relating to changes in gas flow inlet conditions. In many applications, such variances are avoided or compensated for by employing a flow conditioner, such as a screen or a straight tubing section to provide known flow characteristics to the gas flow entering the sensor. However, in respiratory monitoring applications, the exact geometry of the components "upstream" of the sensor ("upstream" being bi-directional, as both inspiration and expiration of the patient are monitored) may vary widely based upon the preference of the clinician and the needs of the patient. In addition, the added volume and resistance to flow resulting from the deployment of a flow conditioner diminish respiratory gas exchange, a particularly undesirable situation with anaesthetized patients.
Differential pressure flowmeters of the prior art employing pressure ports which are flush with the conduit wall, spaced therefrom or facing directly into the gas flow are susceptible to localized pressure effects, Bernoulli effects, and pitot tube effects. Pressure port design in the prior art has failed to minimize such effects and to make prior art flowmeters independent of upstream geometry without adding significant volume to the system and/or substantial resistance to flow.
Localized pressure effects arise in flowmeters when gas flow inlet conditions are asymmetrical, such as occurs when a bend is placed in the flow path in close proximity to the sensor, when a jet or nozzle intrudes on the flow stream, or when any non-symmetrical obstruction is placed in the inlet-stream.
The Bernoulli effect occurs when fluid flow passes over a tube or other structure placed perpendicular to the direction of flow, the flow over the obstruction causing a vacuum which leads to errors in the measurement of differential pressure across an obstruction to the flow.
The pitot tube effect, or "ram" effect, is related to flow velocity, as the port of a pitot tube faces toward the direction of gas flow. When a nozzle or jet is placed upstream of a sensor, a localized high velocity flow is created in the center of the flow stream, leading to erroneous results in devices of the type disclosed in the prior art.
The flow sensor design of the aforementioned '773 patent is susceptible to error from all of the above phenomena, by virtue of the use of a large number of small pressure ports or apertures placed about the cross-section of the housing bore and the placement of such ports facing the flow direction on the leading edges of the supporting ribs. The '773 sensor is also susceptible to clogging and error from mucous and other patient fluids due to the close proximity of some of the ports to the inner wall of the sensor housing.
The flow sensor design of the aforementioned '332 patent is somewhat less susceptible to clogging from patient fluids due to its axial port location, but is very susceptible to localized pressure effects due to the configuration of the leading faces of the vanes or baffles supporting the pressure ports, which structure collects or focuses the gas flow from across the cross section of the sensor housing bore directly into the pressure ports. This configuration also renders the device of the '332 patent very susceptible to error from the pitot tube effect under certain inlet conditions, and has been demonstrated to unduly limit the dynamic range of the device.
In short, all known prior art differential pressure flow sensors suffer deficiencies when exposed to less than ideal gas flow inlet conditions, and further possess inherent design problems with respect to their ability to sense differential pressure in a meaningful, accurate, repeatable manner over a substantial dynamic flow range.